tigercontrol.models.time_series.RNN¶
-
class
tigercontrol.models.time_series.
RNN
[source]¶ Description: Produces outputs from a randomly initialized recurrent neural network.
Methods
__init__
()Initialize self. forecast
(x[, timeline])Description: Forecast values ‘timeline’ timesteps in the future :param x: Value at current time-step :type x: float/numpy.ndarray :param timeline: timeline for forecast :type timeline: int help
()Description: Prints information about this class and its methods. initialize
([n, m, l, h, optimizer, loss, lr])Description: Randomly initialize the RNN. predict
(x[, timeline])Description: Predict next value given observation :param x: Observation :type x: float/numpy.ndarray to_ndarray
(x)Description: If x is a scalar, transform it to a (1, 1) numpy.ndarray; otherwise, leave it unchanged. update
(y)Description: Updates parameters :param y: True value at current time-step :type y: int/numpy.ndarray Attributes
compatibles
spec